4.4 years ago by
United States
Hello,
Map Genes back to the constituent transcripts using the gene/transcript mapping files produced by Cuffdiff. The follow the transcripts back to Cufflinks results where full gtf lines are present. (The tool "Select" and/or "Filter" can be used to subset lines from these files). Use these coordinate data with the tool "Extract Genomic DNA" to obtain a version of the transcript based on genomic content.
Note that any base-level variation present in the NGS reads, not present in the reference genome, that may have contributed to the creation of those transcripts, will not be included. If you are searching for a known transcript (from a reference annotation file provided), that is also a valid source - and you can often obtain the transcript directly using that source's identifier at 3rd party data sources - many available in the tool group "Get Data" (UCSC, BioMart, etc.).
Even if you used a Custom Reference genome for your analysis, this is possible. Help is here:
http://wiki.galaxyproject.org/Support#Tools_on_the_Main_server:_Extract_DNA
More can be done here... if willing to move to a cloud Galaxy, although some tools may function on a local Galaxy with proper management of resources. Testing following close review of tool documentation (requirements) will guide - including a targeted (regional) transcriptome assembly that includes just the NGS reads that mapped into the gene bound(s) of interest (this will reduce the memory footprint). The Tool Shed offers several options for this path under the group "Assembly", and the "Learn" section of our wiki lists examples of usage in the form of tutorials for the more common choices.
Best, Jen, Galaxy team
